Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 26, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Deep learning-based segmentation of human oocytes with cross-dataset evaluation.

Zhilin Lei1, Xiaohu Xu1, Liza Tilia2

  • 1Graduate School of Biomedical Engineering, University of New South Wales, Sydney, 2052, NSW, Australia.

Computer Methods and Programs in Biomedicine
|June 24, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Corrigendum to "The central amygdala as a pathological hub: A multi-level review of pain-emotion comorbidity" [Brain Res. Bull. 236 (2026) 111756].

Brain research bulletin·2026
Same author

Redox responsive nanosystems for medicine: from reactive species mapping to precision treatment.

Advanced drug delivery reviews·2026
Same author

Use of low versus high oxygen tension during pre-maturation only in capacitation in vitro maturation (CAPA-IVM): impact on oocyte maturation and metabolism.

Journal of assisted reproduction and genetics·2026
Same author

Synergistic Co<sub>2</sub>N nanoparticles and N-doped porous ginger-derived carbon as an efficient bifunctional oxygen electrocatalyst for Zn-air battery.

Journal of colloid and interface science·2026
Same author

CbAgo-enriched Cas12a biosensor for cancer mutations screening.

Analytica chimica acta·2026
Same author

Multidimensional data analysis and classification using SMIAL.

Communications biology·2026
This summary is machine-generated.

Deep learning models accurately segmented human oocyte regions, with SegFormer showing superior performance in oocyte quality assessment for assisted reproductive technologies. Task-specific pretraining improved cross-site robustness.

Area of Science:

  • Assisted reproductive technologies
  • Medical image analysis
  • Deep learning in embryology

Background:

  • Accurate segmentation of human oocyte subregions is crucial for objective oocyte quality assessment in assisted reproductive technologies.
  • Current methods lack objectivity and automation, necessitating advanced computational approaches.

Purpose of the Study:

  • To segment four key human oocyte regions: ooplasm, perivitelline space (PVS), zona pellucida (ZP), and first polar body (PBI).
  • To evaluate the performance of deep learning models, including SegFormer, for oocyte subregion segmentation across multiple clinical datasets.
  • To assess the impact of domain adaptation and task-specific pretraining on model robustness.

Main Methods:

  • Four deep learning models (UNet++, DeepLabV3+, SegFormer, transformer-based) were trained on a clinically annotated dataset.
Keywords:
Deep learningMicroscopy imagesMulti-dataset learningOocyte qualityOocyte segmentationReproductive medicine

Related Experiment Videos

Last Updated: Jun 26, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

  • Model performance was evaluated using Intersection over Union (IoU) and Dice Similarity Coefficient (DSC).
  • External datasets from different clinical sites were used for testing, with domain adaptation techniques applied for cross-site validation.

    • Main Results:

    • SegFormer demonstrated superior performance compared to convolutional neural network (CNN) models for oocyte segmentation.
    • The SegFormer model achieved high DSC scores: >99% for ooplasm, 87% for PVS, 94% for ZP, and 86% for PBI on the primary dataset.
    • Task-specific pretraining on oocyte data improved model performance on external clinical images compared to ImageNet-based initialization.

    Conclusions:

    • Deep learning enables accurate multi-component human oocyte segmentation.
    • The study provides a foundation for developing automated systems for oocyte quality assessment in assisted reproductive technologies.